Every one is a real build with a fixed price and a delivery window. Next to each is the arithmetic showing what it has to save or earn before it has paid for itself. Check the maths against your own business, then come and tell me where I am wrong.
Fixed price agreed before anything starts. No hourly, no retainer, no surprise invoice.
Read this before you read the numbers.
The price is what I charge to build it. That part is firm and it does not move after we agree it.
The payback line is arithmetic, not a case study. It takes the assumption printed beside it and divides. If that assumption is wrong for your business then the number is wrong for your business, and I would rather you told me so than believed it. I have no interest in selling you a payback that never arrives.
Seven are marked Running now. That means the exact thing described already exists and runs today, either for a client or in my own work, so the build risk is close to zero. The other twenty-nine I would build from parts I have built before. That is a weaker claim and I am not dressing it up as the first one.
Nothing matches that combination. Try another filter.
All of these attack the same problem: people who would have bought from you never found you, or found you and were never followed up.
Type in any local business. It reads their Google profile and website, scores the gaps, and renders a branded PDF you can send them the same hour.
For: marketing agencies and anyone who sells to local businesses.
Researching one prospect by hand is roughly 3 hours. At 20 prospects a month and $50/hr that is $3,000 of time. Pays for itself in month two.
Sweeps a trade and a radius, scores every business on how badly they need what you sell, flags the fake listings, and hands you a keyboard-driven call queue with a written opener per lead.
For: anyone whose growth is limited by not knowing who to call next.
A bought list of 150 local leads runs $300 to $900 and is stale on arrival. This regenerates on demand. Pays for itself after about five list purchases.
When a job is marked done, the customer gets a text asking for a review with the link already built. Unhappy replies get routed to you privately instead of to Google.
For: trades, clinics, restaurants, anyone judged on a star rating.
Going from 20 reviews to 60 moves you up the local pack. If that brings 2 extra jobs a month at $400, it repays the build in under 3 months.
Any call you do not pick up gets an automatic text within seconds, asking what they need and offering a callback slot. Every reply lands in one place.
For: one-person and small-crew businesses who cannot answer while working.
Most people who reach voicemail call your competitor instead. Recovering 1 job a month at $400 repays the build in about 4 months, then it is profit.
A short guided form that asks the three questions you actually need, prices the obvious cases instantly, and books a visit for the rest. No more phone tag to discover the job is too small.
For: service businesses drowning in unqualified enquiries.
If you spend 6 hours a month qualifying jobs you end up declining, that is $300 of your time. Repays in about 7 months on time alone, sooner on the jobs it wins.
Generates a proper page for every service in every town you cover, each with its own schema markup, instead of one page listing 30 towns that ranks for none of them.
For: any business that serves more than one town.
Agencies charge $150 to $400 per landing page. Thirty pages is $4,500 at the low end. Cheaper than the first batch, and the next thirty are free.
Checks your competitors' public prices and stock on a schedule and emails you only when something moves. No dashboard you have to remember to open.
For: retail, e-commerce, and trades in a crowded market.
Catching one underpriced line before a slow quarter is worth more than the build. On $200k turnover, a 1% margin correction is $2,000 a year.
Taking money on your own site, under your own name, without a platform sitting between you and your customer taking a cut.
Sell files. Customer pays, gets an automatic email with a secure download link, and their own account to re-download later. Card and PayPal both work.
For: anyone selling software, templates, courses, plans, music or PDFs.
A marketplace keeps 30% to 50%. On $2,000 a month of sales that is $600 to $1,000 gone. Your own store repays the build in 5 to 8 months and keeps paying.
One QR code that opens your profile, saves your contact straight to their phone, and shows your links. Update it any time without reprinting anything.
For: sales teams, trades, realtors, anyone who hands out cards at events.
Printing 500 cards is $80 and they are wrong the moment a number changes. For a team of 10 reprinting twice a year, this removes about $1,600 a year of print.
Customers pick a slot from your real availability and pay a deposit to hold it. Confirmation and reminder texts go out on their own.
For: clinics, salons, tradespeople, consultants, anyone with a calendar and no-shows.
No-shows typically run 10% to 20%. A deposit cuts that hard. On 40 bookings a month at $120, halving a 15% no-show rate recovers $360 a month.
Recurring billing, member-only content, automatic dunning when a card fails, and a page where members manage themselves instead of emailing you.
For: anyone turning a one-off product into monthly revenue.
Failed-card dunning alone usually recovers 20% to 40% of involuntary churn. On 100 members at $30, that is $180 to $360 a month you were losing silently.
A link the customer can check any time to see exactly where their order or job is, instead of calling you to ask.
For: custom manufacturing, repairs, print shops, anyone with a lead time.
If "where is my order" takes 5 hours a month at $25/hr, that is $125 a month. Repays in about 19 months on time alone, and it is really a complaints fix.
Someone fills a cart and leaves. Three messages go out on a schedule you set, the last one with a discount you control.
For: any store already taking orders.
Recovery sequences typically bring back 5% to 10% of abandoned carts. On $5,000 a month of abandonment that is $250 to $500 a month. Repays in 4 to 7 months.
Other sellers list their own products on your site. You take a cut of every sale automatically, and they manage their own listings and payouts.
For: trade associations, local co-ops, anyone with a community and no product of their own.
At a 10% cut, $10,000 a month of vendor sales is $1,000 a month to you for inventory you never buy. Repays in about a year at that volume.
A booking page for the chargers you already own. Drivers reserve a slot, pay by card, and you get a monthly statement showing who used what. Works with the hardware you have.
For: hotels, gyms, restaurants and small property owners with two to eight chargers.
Four chargers at three paid sessions a day at $8 a session is about $2,880 a month. Most small sites lose a chunk of that to drivers who cannot tell whether a bay is free. Recovering a third of those sessions pays for the build in month two.
The least glamorous category and usually the highest return, because those hours are already being spent and nobody is billing anyone for them.
One record moves from quote to accepted to scheduled to invoiced to paid. Nothing is retyped, and nothing sits in the accepted column with no invoice against it.
For: any business that quotes before it bills.
Retyping a quote into an invoice is 10 minutes. At 60 jobs a month that is 10 hours. At $50/hr that is $500 a month, repaying in about 7 months, before counting invoices that never got sent.
Crew takes before and after photos on their phone. A branded PDF report is waiting in the customer's inbox before the van has left the street.
For: restoration, roofing, pest control, landscaping, anything an insurer or landlord needs proof of.
Assembling one report by hand is 20 to 30 minutes. At 40 jobs a month that is 16 hours. At $30/hr that is $480 a month, repaying in under 6 months.
Staff clock in from their phone with location. At period end it produces the exact file your payroll provider wants, with overtime already worked out.
For: any business with hourly staff and a spreadsheet.
Manual payroll prep is commonly 4 to 6 hours a period. At 24 periods and $40/hr that is $3,840 to $5,760 a year, so it repays inside the first year and ends the arguments about hours.
Forward a supplier invoice, a receipt or a signed form to one address. The numbers come out as clean rows in your system, with anything uncertain flagged for a human rather than guessed.
For: bookkeepers, property managers, clinics, anyone retyping paper.
Keying 300 documents a month at 3 minutes each is 15 hours. At $30/hr that is $450 a month, repaying in about 11 months, and the error rate drops at the same time.
Watches stock against how fast each line actually moves and tells you what to order this week, before you find out by running out.
For: retail, workshops, clinics, kitchens.
One stockout on a fast line costs a weekend of sales. If that is $800 and it happens three times a year, the build repays in year one on prevented stockouts alone.
Every contract customer gets rebooked automatically on their own cycle, routed sensibly by geography, with the reminder already sent.
For: maintenance contracts, cleaning, inspections, lawn care, HVAC servicing.
Contracts that lapse because nobody rebooked them are the quietest revenue loss there is. Saving 5 contracts a year at $600 is $3,000, repaying inside 16 months and compounding after.
Not a chatbot bolted onto your website. These are jobs a person does today that a model can do with a human checking the output, built so that when it is unsure it says so instead of inventing.
Ask questions of your own manuals, contracts, price lists and past jobs, and get an answer that cites the document it came from. If the answer is not in your documents, it says so.
For: firms where one or two people are the only ones who know things.
If five staff each lose 30 minutes a day hunting for information, that is 10 hours a week. At $35/hr that is $1,400 a month, repaying in about 5 months.
Reads the shared mailbox, sorts what is a new enquiry, a complaint, a supplier or noise, drafts the routine replies, and leaves the rest for you with the context already gathered.
For: any business where a shared inbox is the real bottleneck.
Two hours a day of triage at $35/hr is $1,540 a month. Removing even half of that repays the build in under 6 months.
Recorded calls become a written summary, the agreed next step, and a dated follow-up task in your CRM. Nobody types notes after a call ever again.
For: sales teams and anyone whose CRM is full of blank activity fields.
Post-call notes are 5 minutes a call. At 25 calls a day across 3 people that is 31 hours a month. At $40/hr that is $1,250 a month, repaying in 3 months.
Feed it the notes from a discovery call plus your past winning proposals. It returns a first draft in your format with your pricing, ready to edit rather than ready to send.
For: consultants, agencies, contractors bidding regularly.
A serious proposal is 3 to 4 hours. At 8 a month and $75/hr that is $1,800 to $2,400. Cutting each to one hour repays the build in about 2 months.
A test suite for the AI features you already shipped. It proves they still behave after every change, and it is built so that it can actually fail. A suite that cannot go red is not a suite.
For: companies who have put a model in front of customers and cannot tell when it drifts.
One confidently wrong answer to a customer costs more than this build. It is insurance, priced at roughly one bad incident.
Takes the messy export, matches duplicates, fixes formats, flags what it is not sure about, and never silently deletes a row. Everything it changed is listed for you.
For: anyone merging lists, migrating systems, or living in a CRM full of duplicates.
A one-off cleanup quoted out is $1,500 to $4,000 and you pay it again next year. This repays on the second cleanup and then runs whenever you want.
Lets Claude or ChatGPT safely read and act on your own systems through a small set of tools you define, with hard limits on what it may touch and what it may say.
For: firms who want AI on their real data without handing over the keys.
The alternative is staff pasting company data into a public chat window, which is a breach waiting to be discovered. Priced well below what that costs to explain.
The headset is no longer the hard part. These all run in a browser on a phone as well as in a headset, so your customer does not have to own hardware to see the value.
Customers walk through your full product range at real scale, in a headset or straight in the browser on their phone. No warehouse, no floor space, no travel.
For: furniture, kitchens, cars, boats, machinery, modular buildings.
A physical showroom is rent, staff and stock. If it replaces one 400 sq ft showroom at $2,200 a month, it repays in under 5 months and it is open at 2am to a buyer in another state.
Staff practise the dangerous procedure without the danger, as many times as they need, and every attempt is scored and recorded for your compliance file.
For: construction, manufacturing, utilities, warehousing, anyone with a safety record to defend.
One recordable injury costs an employer tens of thousands before insurance moves. Taking a single incident out of a year pays for this several times over, and the training record is the evidence.
Customer points their phone at the room and sees your product sitting in it at the right size, in the right colour, before they order.
For: furniture, paint, flooring, signage, blinds, appliances.
Returns on furniture and decor commonly run 15% to 25%, and "it did not fit the space" is the top reason. On $120k of annual sales, cutting returns by a third keeps roughly $6,000 of margin.
A walkthrough of your property, venue or workshop with clickable hotspots that explain what the visitor is looking at. Works on any phone, no app to install.
For: real estate, hotels, venues, schools, care homes, gyms.
Photographers charge $400 to $900 per property tour. At six properties a year you are past the build cost, and you own the tool rather than rebooking the shoot.
A new hire or visiting contractor walks your actual site in VR before they ever set foot on it, and arrives knowing the layout, the hazards and the rules.
For: anyone inducting contractors, seasonal staff or high-turnover crews.
A supervisor walking each new starter round is 2 hours. At 60 starters a year and $45/hr that is $5,400 of supervisor time, repaying in under two years and the induction never varies by who gave it.
Smaller, faster, and priced by the job. You already own the software or the site. You just need someone to make it work properly and then get out of your way.
You bought the software and it is sitting in a folder. I install it on your server, configure it, connect your payment and email, test it end to end with a real transaction, and show you how to run it.
For: anyone holding a licence, a bundle or a plugin they have never managed to get working.
Software you own and cannot run is worth nothing. If the bundle cost you $500 and it is still in a folder, this is the cheapest way to make that $500 real.
Move a site to hosting that actually works, or recover one that broke, without losing your products, orders, customers or search rankings. Full backup taken before anything is touched.
For: anyone on hosting that keeps falling over, or who has lost files and does not know why.
A store that is down or broken is making nothing. If you take $150 a day online, eleven days of downtime costs more than this job.
Get card and PayPal payments working properly on the site you already have, including the receipt email, the refund path and a real test transaction before you go live.
For: anyone taking payment by bank transfer, cash, or a link they paste into a message.
Card checkout converts far better than "message me to pay". Two extra sales a month at $60 repays this inside a year, and you stop chasing transfers.
Hundreds or thousands of products onto your store in one pass, with images, prices, categories, descriptions and download files, generated from whatever list you already have.
For: anyone facing a catalogue too big to type in by hand.
Adding a product by hand is 4 to 6 minutes. For 500 products that is 40 hours, so at $25/hr you are $1,000 ahead and it can be re-run when prices change.
Not included on purpose: ongoing retainers, app store mobile apps, and anything where I would end up the only person who understands it afterwards. If your idea is not on this list, ask anyway. If it is a bad fit I will say so, and it costs you one message.
Send it and it lands in my inbox. I read every one myself and reply, usually the same day. If the payback maths does not hold for your business I will tell you that rather than quote you.
Prefer to talk instead of type? Any of these reaches me directly.
It is in my inbox. I will read it myself and reply, usually the same day.
In a hurry? WhatsApp or 760-481-4120.